A method and system for path planning of equipment groups without human intervention based on RRT* algorithm

Abstract

The present invention relates to the field of intelligent construction of civil engineering, and specifically to a method and system for path planning of a group of unmanned equipment based on the RRT* algorithm. The method comprises the following steps: obtaining information on the number, starting point, end point, travel length, task urgency, energy efficiency and obstacle area of ​​unmanned equipment; determining the priority ranking of the equipment in path planning based on the information; determining the shortest path of the highest priority equipment under multiple obstacle conditions based on the priority ranking; performing collision checks on the travel paths and rotation paths of multiple equipment based on the shortest path; and realizing multi-dimensional obstacle avoidance global path collaborative planning of the equipment group through the results of the collision check. The present invention reduces the equipment travel time and improves the overall construction efficiency by determining the equipment priority; and avoids the limitation of only considering travel obstacle avoidance by integrating the travel rotation obstacle avoidance strategy and the static-dynamic obstacle avoidance strategy, thereby improving the applicability and accuracy of path planning.

Description

Technical Field

[0001] The present invention relates to the field of intelligent construction of civil engineering, and in particular to a method and system for planning a path for a group of equipment without human intervention based on an RRT* algorithm. Background Art

[0002] The rapid development of information and intelligent technologies is profoundly transforming how society operates. Leveraging artificial intelligence to optimize the working environment and improve construction efficiency in traditional civil engineering has become a crucial path to driving the industry's transformation and upgrading, and towards sustainable development. Especially in engineering projects with extremely complex working environments, such as those at high altitudes and deep burial depths, the construction of intelligent, unmanned sites is of immeasurable importance for ensuring construction safety and quality. Multi-machine collaborative path planning, as the core component of achieving unmanned sites, utilizes advanced intelligent algorithms to optimize the travel paths and operating sequences of various types of machinery and equipment. This not only significantly improves construction efficiency, effectively reduces conflicts and waiting times between equipment, but also achieves optimal resource allocation. This collaborative operation model enables seamless integration of different types of machinery and equipment, improving operational continuity and overall efficiency.

[0003] In the field of path planning, a variety of algorithms have been proposed and widely used. For example, the rapidly expanding random tree algorithm (RRT algorithm), based on the concept of random sampling, is a heuristic algorithm widely used in path planning problems. The RRT algorithm randomly samples points in space and expands these points onto a path tree, allowing a moving device to gradually generate a path from its starting point to its destination. However, the path generated by the RRT algorithm is often not the optimal path. To address this issue, an improved version of the RRT algorithm, the RRT* algorithm, was developed. The RRT* algorithm introduces a path optimization mechanism based on the RRT algorithm, which can generate shorter and more optimal paths by reconnecting path nodes and parent nodes. Guided by this algorithm, a device can quickly and efficiently plan the shortest path in an environment with dense obstacles and complete the task. The above path planning methods have significant advantages in single-device walking obstacle avoidance, but still have shortcomings in multi-device coordination and rotation obstacle avoidance. First, the task priority, energy efficiency, mobility and environmental factors of the equipment are not taken into consideration, which increases the complexity of the post-equipment planning path in the multi-machine collaborative planning process, further prolongs the equipment transportation time and significantly reduces construction efficiency; second, a simple exclusion principle is usually adopted to avoid static collisions, but it is difficult to control the real-time interaction and rotation posture of the equipment in a dynamic and unmanned environment; third, only obstacle avoidance while moving is considered, and rotation obstacle avoidance is not considered, which poses a greater risk of equipment collision.

[0004] At present, there is not much research on the global collaborative path planning of multi-device obstacle avoidance without human intervention. There is no specific path planning method that simultaneously considers the priority planning order of multiple devices, collaborative dynamic collision detection and rotation obstacle avoidance. Summary of the Invention

[0005] In view of the defects in the prior art, the present invention provides a method and system for human-free equipment group path planning based on the RRT* algorithm.

[0006] On the one hand, the present invention provides a method for path planning of a group of unmanned equipment based on the RRT* algorithm, including the following steps: obtaining information on the number, starting point, end point, travel length, task urgency, energy efficiency and obstacle area of ​​the unmanned equipment; determining the priority ranking of the equipment in path planning based on the information; determining the shortest path of the highest priority equipment under multiple obstacle conditions based on the priority ranking; performing collision checks on the travel paths and rotation paths of multiple equipment based on the shortest path; and realizing global collaborative path planning of multi-dimensional obstacle avoidance for a group of equipment through the results of the collision check. The present invention is based on the RRT* algorithm to realize the global collaborative path planning of multi-dimensional obstacle avoidance of equipment groups without manpower, and has achieved the following effects: First, it comprehensively integrates the number, location, task requirements and energy consumption information of equipment without manpower, and ensures priority planning of paths for emergency task equipment through a priority sorting mechanism, thereby improving construction efficiency and task response speed; second, in a complex multi-obstacle environment, it accurately calculates the shortest path of the highest priority equipment, effectively shortens the equipment travel time, and reduces energy consumption at the same time, thereby achieving efficient and energy-saving path planning; third, through detailed collision checks on the travel and rotation paths of multiple equipment, the equipment path is dynamically adjusted, thereby realizing multi-dimensional obstacle avoidance and collaborative planning of the equipment group on a global scale, which not only avoids conflicts and collisions between equipment, but also improves the operating safety and efficiency of the entire construction site equipment group, providing solid technical support for the efficient operation of unmanned sites.

[0007] Optionally, determining the priority ranking of devices in path planning based on the information includes: constructing a priority evaluation model for devices in path planning based on the information; and determining the priority ranking of devices in path planning using the priority evaluation model. The present invention systematically integrates and analyzes various types of relevant information, such as the energy consumption characteristics of the equipment, the urgency of the task, and the complexity of the path, by constructing a priority evaluation model for devices in path planning, providing a scientific basis for subsequent priority sorting, ensuring the comprehensiveness and accuracy of the evaluation, and avoiding subjective assumptions and one-sidedness; using the priority evaluation model for sorting to quickly and objectively determine the priority of the equipment in path planning, greatly improving decision-making efficiency and accuracy, helping to optimize resource allocation, ensuring that key equipment and urgent tasks are given priority, and thus improving overall operational efficiency; dynamically adjusting priority sorting according to the needs of different devices and tasks to cope with various complex and changing scenarios, and ensuring the rationality and effectiveness of path planning.

[0008] Optionally, the priority evaluation model satisfies the following expression:

[0009] Q i =αP i +βE i +γT i

[0010] Among them, Q i is the priority evaluation score of the i-th device, P i is the task urgency of the i-th device, E i is the energy efficiency of the i-th device, T i is the path obstacle impact of the i-th device, and α, β, and γ are weight coefficients. The priority evaluation model constructed by the present invention has the following effects: First, it comprehensively considers three key factors: the urgency of the device task, energy efficiency, and the impact of path obstacles, fully reflecting the actual needs and limitations of the device in path planning, ensuring the objectivity and accuracy of the evaluation results; second, by introducing weight coefficients, the weights of various factors can be flexibly adjusted according to different application scenarios and needs, thereby achieving personalized evaluation of the priorities of different devices, enhancing the applicability and flexibility of the model; third, the use of a simple linear combination method is simple and fast to calculate, and the priority evaluation score of the device is quickly obtained, providing strong support for rapid decision-making in path planning, effectively improving the efficiency and quality of path planning.

[0011] Optionally, the method of determining the shortest path for the highest priority device under multiple obstacle conditions based on the priority sorting includes: determining the highest priority device based on the priority sorting; planning the travel path of the highest priority device to determine the shortest path under multiple obstacle conditions, wherein the planning includes travel collision checking and rotation collision checking. The present invention quickly locks the highest priority device through priority sorting, ensuring that critical or emergency mission equipment obtains path planning services first, significantly improving the efficiency of resource allocation and task execution; conducting detailed travel path planning for the highest priority device, and comprehensively eliminating potential obstacles in the path through travel collision checking and rotation collision checking, ensuring that the device moves safely and smoothly in a complex multi-obstacle environment, and reducing the risk of device damage and mission failure; comprehensively considering the feasibility and optimality of the path, while meeting safety requirements, it also ensures that the device can reach the destination by the shortest path, further improving the efficiency and effectiveness of the overall task execution.

[0012] Optionally, the rotational collision check includes the following steps: preliminarily determining the collision range based on the spatial intersection between the obstacle and the rotation range of the highest priority device; determining the rotation path vector based on the collision range; and judging the rotation direction and rotational collision risk based on the rotation path vector. The present invention preliminarily determines the range in which the highest priority device may collide with obstacles during rotation through analysis of spatial intersection, providing key information for subsequent path planning and collision risk assessment, ensuring the accuracy and safety of the planning process; further determining the rotation path vector based on the collision range not only clarifies the specific movement direction and path of the device during rotation, but also provides a basis for judging the rotation direction and collision risk, enhancing the operability and practicality of path planning; accurately judging the rotation direction of the device and the potential rotational collision risk through analysis of the rotation path vector, providing a strong guarantee for the safe and efficient movement of the device in a complex environment, and effectively improving the intelligence level of path planning and the efficiency of task execution.

[0013] Optionally, the method of using the vector cross product method to judge the rotation direction and rotation collision risk based on the rotation path vector includes: judging the counterclockwise rotation collision risk based on the rotation path vector; judging the clockwise rotation collision risk based on the rotation path vector. The present invention accurately judges the collision risk that the device may encounter when rotating counterclockwise through the analysis of the rotation path vector, providing an important guarantee for the safe movement of the device in a complex environment, and effectively avoiding task delays or equipment damage caused by collisions; similarly, based on the rotation path vector, the device accurately assesses the collision risk when rotating clockwise, ensuring the all-round safety of the device during the rotation process, and further improving the intelligence and refinement of path planning; using the mathematical principle of the vector cross product method, a fast and accurate judgment of the rotation direction and collision risk is achieved, providing strong support for the real-time path adjustment and dynamic obstacle avoidance of the device, and significantly improving the efficiency and flexibility of path planning.

[0014] Optionally, the planning of the travel path of the highest priority device and determining the shortest path under multiple obstacle conditions includes: generating a root node of the path tree at the starting point of the highest priority device; randomly sampling within the travel area of ​​the highest priority device to generate a random point; determining the shortest distance from the random point to each node of the path tree as the parent node; generating a new path at the parent node in the direction toward the random point with a length equal to the travel length, and the end of the new path is the proposed growth node of the path tree; comparing the distance between the new path and the obstacle with the size of the traveling device to determine whether there is a collision risk; adding the proposed growth node to the path tree based on the result of no collision risk; identifying existing nodes within a certain range of the proposed growth node, and taking the point with the shortest total path from the starting point to the proposed growth node as the new parent node; determining whether the total path distance of the existing nodes after reconnection with the proposed growth node as the parent node is shortened, and determining whether there is a collision risk; performing a threshold judgment on the distance between the proposed growth node and the end point to determine the shortest path under multiple obstacle conditions. The present invention generates a path tree and randomly samples random points to flexibly explore possible paths in a complex environment, effectively avoiding the local optimal solution problem that traditional path planning methods may fall into, and improving the globality and optimality of path planning; through the generation of parent nodes and new paths, and the judgment of collision risks, the safety and feasibility of the equipment during travel are ensured, path interruption or task failure caused by collision is avoided, and the reliability and stability of path planning are improved; by identifying existing nodes within a certain range of the proposed growth node and judging whether the total path distance after reconnection is shortened, the path tree is continuously optimized to find the shortest path from the starting point to the end point, thereby improving the efficiency and quality of path planning and providing strong support for the rapid and accurate movement of the equipment.

[0015] Optionally, the collision check of the travel paths and rotation paths of multiple devices based on the shortest path includes: based on the shortest path of the highest priority device, performing a collision check on the travel paths of low-priority devices; based on the shortest path of the highest priority device, performing a collision check on the travel and rotation paths of low-priority devices. The present invention ensures that all devices do not interfere with or collide with each other during travel by performing a collision check on the travel paths of low-priority devices, thereby ensuring a safe distance and smooth movement between devices; performing a collision check on the travel and rotation paths of low-priority devices not only considers the linear movement of the device, but also takes into account the rotational movement of the device, thereby comprehensively improving the safety and reliability of path planning; the collision check mechanism effectively avoids path adjustments or task delays caused by conflicts between devices, improves the efficiency of overall path planning and the smoothness of task execution, and provides a strong guarantee for the collaborative operation of multiple devices.

[0016] Optionally, the collision check includes: the low-priority device performs a collision check on the travel path according to the paths planned by all higher-priority devices; the low-priority device performs a collision check on the rotation and travel path according to the paths planned by all higher-priority devices. The low-priority device of the present invention comprehensively considers the paths planned by all higher-priority devices, performs a collision check on the travel path, ensures the orderly travel between devices, avoids collision accidents caused by path overlap or conflict, and improves the safety and overall efficiency of the work area; the collision check mechanism not only focuses on the straight-line travel path of the equipment, but also conducts in-depth collision checks on the rotation and travel paths, fully considering the dynamic changes of the equipment during the operation process, and further enhancing the safety and adaptability of path planning; through comprehensive collision checks, low-priority devices can plan their own paths more flexibly while ensuring safety, effectively reducing the time loss caused by path adjustment, and improving the consistency and efficiency of the overall work process.

[0017] In a second aspect, the present invention provides a system for planning a path for a group of unmanned equipment based on an RRT* algorithm, comprising an input device, a processor, an output device, and a memory, wherein the input device, the processor, the output device, and the memory are interconnected, wherein the memory is used to store a computer program, the computer program including program instructions, the processor is configured to call the program instructions, and the system uses the method for planning a path for a group of unmanned equipment based on an RRT* algorithm. The system for planning a path for a group of unmanned equipment based on an RRT* algorithm provided by the present invention is highly integrated and has the following effects: First, through the efficient search capability of the RRT* algorithm, it can quickly plan a safe and efficient path for a group of equipment in a complex construction site environment, effectively avoiding collisions and path conflicts between equipment, and ensuring the orderly progress of construction site operations; second, the optimization capability of the RRT* algorithm makes the planned path smoother and more continuous, reducing the number of sudden stops and turns of equipment, thereby improving the operating efficiency and lifespan of the equipment; third, the system also has strong adaptability and flexibility, and can dynamically adjust and optimize according to different construction site environments and equipment requirements, realizing adaptive and intelligent path planning.

[0018] Compared with the existing technology, the beneficial effects of the present invention include: by determining the priority of the equipment in path planning, the equipment travel time is reduced and the overall construction efficiency is improved; it integrates the travel-rotation obstacle avoidance strategy and the static-dynamic obstacle avoidance strategy, optimizes the path planning of the equipment in a complex construction site environment, avoids the limitations of only considering travel obstacle avoidance in traditional path algorithms, and improves the applicability and accuracy of path planning; it is particularly suitable for unmanned land scenarios, makes up for the defects of traditional planning methods that are difficult to adapt to multiple obstacles and multiple devices, and provides reliable technical support for the collaborative work of intelligent and unmanned construction equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 Flowchart of a method for human-free device group path planning based on the RRT* algorithm according to an embodiment of the present invention;

[0020] Figure 2 This is a schematic diagram of a rotation direction vector considered when the device according to an embodiment of the present invention rotates;

[0021] Figure 3 This is a schematic diagram of the intersection of the travel paths of the front-end device and the rear-end device according to an embodiment of the present invention;

[0022] Figure 4 A schematic diagram illustrating the intersection of the travel and rotation paths of the front device and the rear device according to an embodiment of the present invention;

[0023] Figure 5 This is a schematic diagram of the global path collaborative planning results of a device group multi-element obstacle avoidance according to an embodiment of the present invention;

[0024] Figure 6 Schematic diagram of the structure of the non-manual equipment group path planning system based on the RRT* algorithm according to an embodiment of the present invention. DETAILED DESCRIPTION

[0025] Specific embodiments of the present invention will be described in detail below. It should be noted that the embodiments described herein are for illustrative purposes only and are not intended to limit the present invention. In the following description, numerous specific details are set forth to provide a thorough understanding of the present invention. However, it will be apparent to one of ordinary skill in the art that these specific details are not necessarily required to practice the present invention. In other instances, well-known circuits, software, or methods are not specifically described to avoid obscuring the present invention.

[0026] Throughout this specification, references to "one embodiment," "an embodiment," "an example," or "an example" mean that a particular feature, structure, or characteristic described in connection with the embodiment or example is included in at least one embodiment of the present invention. Therefore, appearances of the phrases "in one embodiment," "in an embodiment," "an example," or "an example" in various places throughout this specification are not necessarily all referring to the same embodiment or example. Furthermore, the particular features, structures, or characteristics may be combined in any suitable combinations and / or subcombinations in one or more embodiments or examples. Furthermore, those of ordinary skill in the art will appreciate that the figures provided herein are for illustrative purposes only and are not necessarily drawn to scale.

[0027] See Figure 1 The embodiment of the present invention provides a method for planning a path for a group of devices without human intervention based on the RRT* algorithm. The method comprises the following steps:

[0028] S1. Obtain information on the number, starting point, end point, travel length, task urgency, energy efficiency, and obstacle areas of unmanned equipment to be utilized.

[0029] S1 includes the following steps:

[0030] S11. Obtain the number of devices to be utilized without human intervention.

[0031] In one embodiment, an equipment list is first collected from the construction site management department or the equipment supplier, and the equipment information on the list is carefully checked to ensure the accuracy and completeness of the information.

[0032] Furthermore, for devices that have not yet been connected to the Internet of Things, it is necessary to install an Internet of Things communication module (such as NB-IoT, LoRa, etc.) or an RFID tag, and configure the network connection to ensure that the device can send data to the remote management system in real time or at a scheduled time.

[0033] Furthermore, an account is created for each device in the remote management system, and the serial number or RFID tag of the device is bound, and monitoring parameters are set so that the status of the device can be monitored in real time.

[0034] Furthermore, the remote management system monitors the status of the equipment in real time, including online status, working status, etc., and counts the number of equipment on the current construction site based on the online status of the equipment.

[0035] Furthermore, to ensure the accuracy of the statistical results, it is necessary to verify the statistical results through on-site inspections, video surveillance, etc. If it is found that the statistical results deviate from the actual situation, it is necessary to adjust the statistical methods of the remote management system or optimize the Internet of Things connection.

[0036] Furthermore, a device quantity report is generated based on the statistical results, wherein the quantity report includes the number of devices that are not working, the number of devices that are working, and the total number of devices.

[0037] Furthermore, the number of devices to be utilized is determined based on the device quantity report.

[0038] S12. Obtain the starting point and end point of the unmanned equipment.

[0039] In one embodiment, first, it is evaluated whether the current location of the equipment is suitable as a starting point, taking into account factors such as its distance to the working area and the complexity of the terrain.

[0040] Furthermore, if the current location is not suitable, one or two alternative starting points are selected based on the site layout and equipment capabilities.

[0041] Furthermore, the accessibility, safety and impact on operational efficiency of alternative starting points are evaluated.

[0042] Furthermore, the most appropriate starting point is determined by comprehensively considering equipment capabilities, site environment, and operational requirements.

[0043] In another embodiment, first, the operating area of ​​the equipment to be utilized is determined.

[0044] Furthermore, one or two alternative endpoints are selected based on the layout of the work area.

[0045] Furthermore, evaluate whether the location of the alternative end point is convenient for the equipment to unload materials, whether it affects the operation of other equipment, and whether it is convenient for the equipment to return to the starting point or the next operating area.

[0046] Furthermore, the most appropriate endpoint is determined by comprehensively considering the operational requirements, equipment capabilities, and site environment.

[0047] S13. Obtain the progress of the equipment without human intervention.

[0048] In one embodiment, first, the technical documentation of the device is consulted to understand key parameters; and device performance testing is performed in a safe environment to verify the parameters and identify possible deviations.

[0049] Furthermore, use surveying equipment to conduct detailed surveying of the construction site, record terrain and obstacle information, and evaluate the adaptability of the equipment under different terrain conditions.

[0050] Furthermore, the work tasks that the equipment needs to complete and the task connection relationship are analyzed; based on the equipment capabilities and the construction site environment, the maximum safe travel length is calculated; according to the work task requirements, the travel length is adjusted to ensure efficient and safe operation.

[0051] Furthermore, the set step length is verified in a simulation environment or an actual construction site; and necessary fine-tuning of the step length is performed based on the verification results.

[0052] S14. Obtain the mission urgency of the unmanned equipment.

[0053] In one embodiment, first, all tasks that need to be performed by unmanned equipment are listed, including task names, descriptions, expected completion times, etc.; and the project manager is communicated with to ensure the completeness and accuracy of the task list.

[0054] Furthermore, each task is assigned a priority level, such as high, medium, or low, based on its urgency and importance. Urgency considerations may include the task's deadline, impact on subsequent work, etc. Importance may be based on the task's impact on the overall progress or quality of the project.

[0055] Furthermore, tasks are assigned to the most suitable unmanned equipment based on its capabilities and characteristics, taking into account the equipment's current status, maintenance schedule, and availability, ensuring that the assigned tasks do not exceed the equipment's carrying capacity.

[0056] Furthermore, the urgency score of each task is calculated by comprehensively considering the priority of the task, the capability of the equipment, and the dependencies between tasks. A simple weighted summation method is used to convert factors such as the priority of the task, the capability of the equipment, and the dependencies into a specific urgency score.

[0057] S15. Obtain energy efficiency of the equipment without human intervention.

[0058] In one embodiment, first, the unmanned equipment that needs to be monitored is determined, and corresponding energy consumption monitoring equipment is prepared to ensure that the monitoring equipment can accurately record the energy consumption data of the equipment and associate it with the operating time and working mode information of the equipment.

[0059] Furthermore, during normal operation, the device's energy consumption data is continuously recorded. This includes the device's total energy consumption, peak energy consumption, and average energy consumption. The device's operating time, operating mode, and environmental conditions are also recorded for subsequent analysis.

[0060] Furthermore, the effectiveness of the equipment in performing various tasks is evaluated, including the amount of tasks completed, operating efficiency, and quality. This can be done through on-site observation, work logs, and task completion reports.

[0061] Furthermore, the energy consumption data and task effectiveness evaluation results are combined to calculate the energy efficiency of the equipment.

[0062] S16. Obtain information about the obstacle area.

[0063] In one embodiment, first, a preliminary scan and record of the obstacle area is performed through drone or ground survey, including basic information such as the location, size, and type of the obstacle.

[0064] Furthermore, tools such as laser scanners or GPS locators are used to conduct detailed mapping of the obstacle area to generate high-precision terrain maps and obstacle distribution maps.

[0065] Furthermore, all data from the survey process are collated to obtain complete obstacle area data information.

[0066] S2. Determine the priority ranking of the devices in the path planning based on the information.

[0067] Among them, S2 includes the following steps:

[0068] S21. Based on the information, a priority evaluation model for the device in path planning is constructed.

[0069] In one embodiment, a priority evaluation model for devices in path planning is constructed by simultaneously considering the mobility performance, task priority, energy consumption characteristics, and environmental impact of the device.

[0070] Specifically, the priority evaluation model satisfies the following expression:

[0071] Q i =αP i +βE i +γT i

[0072] Among them, Q i is the priority evaluation score of the i-th device, P i is the task urgency of the i-th device, E i is the energy efficiency of the i-th device, α, β, γ are weight coefficients; T iis the path obstacle influence of the i-th device, which satisfies the following relationship:

[0073]

[0074] Among them, M i is the straight-line distance from the starting point to the end point of the i-th device, K i is the total number of obstacles on the straight path from the starting point to the end point of the i-th device, G is the total number of obstacles in the entire unmanned area, R i is the progress of the i-th device.

[0075] It should be noted that when there are many obstacles on the straight path from the starting point to the end point of the device, T i will become larger, which will further increase Q i As the value gets bigger, the priority is increased, which means that the more obstacles there are on the shortest path, the easier it is to prioritize them in path planning. This ensures that lower-priority devices with relatively smooth paths will be more flexible in subsequent path planning.

[0076] S22. Determine the priority ranking of devices in path planning using the priority evaluation model.

[0077] In one embodiment, first, a priority evaluation model is used to obtain a priority evaluation score of a device.

[0078] Furthermore, the priority ranking of the devices in the path planning is determined according to the magnitude of the priority evaluation score.

[0079] Specifically, the higher the device's priority evaluation score, the higher its priority in path planning, which means it will be arranged first. The specific relationship between high and low priority devices is expressed as follows:

[0080] There are n devices, with priority order N1, N2, N3...N i-1 、N i , define V j ={N1, N2, N3...N i-1}, V j N i The high priority is i=2,3,4…n, j=1,2,3…n-1.

[0081] S3. Based on the priority ranking, determine the shortest path for the highest priority device under multiple obstacle conditions. S3 further includes the following steps:

[0082] S31. Based on the priority ranking, determine the highest priority device.

[0083] In one embodiment, based on the priority ranking determined in step S22 , the device with the highest priority ranking is selected, also referred to as the highest priority device.

[0084] Specifically, the determination condition for the highest priority device is described by a mathematical relationship, which is as follows:

[0085]

[0086] Among them, N i , N j All are priority sorting, i=1,2,3…n, j=1,2,3…n, i≠j. When this judgment condition is met, the priority sorting is N i The device with the highest priority is the device.

[0087] S32. Plan the travel path of the highest priority device to determine the shortest path under multiple obstacle conditions, wherein the planning includes travel collision checking and rotation collision checking.

[0088] In one embodiment, the shortest path for a single device is planned based on the idea of ​​RRT* random sampling, which mainly includes the steps of "path tree initialization - random sampling - parent node selection - tree expansion attempt - travel collision check - adding new nodes - parent node reselection - node reconnection - target determination - loop until reaching the target".

[0089] Specifically, first, the path tree is initialized, including: defining the starting point x of the highest priority device initial , end point x final , perform step size L and obstacle area; generate the root node of the path tree at the starting point of the highest priority device.

[0090] Furthermore, random sampling is performed. This includes: random sampling within the equipment driving area, generating a random point x rand ; Among them, the sampling range is determined by the driving space of the equipment and the location of obstacles.

[0091] Furthermore, the parent node is selected. This includes: calculating the distance from the random point to each node in the existing path tree, and finding the node closest to it as the parent node x near .

[0092] Further, try to expand the tree. Including: from x near Along the direction x rand A new path to be confirmed with a distance of L is generated in the direction of the path, and the end of the path is the proposed growth node x of the path tree. new .

[0093] Furthermore, the collision check is performed. This includes:near →x new ) performs a collision check on the path, comparing the distance between the path segment and the obstacle with the size of the walking device to determine whether there is a collision risk.

[0094] Furthermore, a new node is added. This includes: if the generated new path segment passes the walking collision check, that is, there is no collision risk, then the new node x new Added to the path tree, its parent node is x near .

[0095] Further, a new parent node is selected. This includes: identifying x new The existing nodes within a certain range (circle with radius R) so that initial to x new The point with the shortest total route distance is x new The new parent node x short Then use the collision check step to determine the new path segment (x short →x new ) Is there a risk of collision? If not, determine x short is x new The new parent node of .

[0096] Furthermore, the nodes are reconnected. This includes: new The existing nodes within a certain range (circle with radius R) are new Try to reconnect the parent node. The connection basis is to judge whether the total path distance is shortened after reconnection. If the distance is shortened and there is no risk of collision, the parent node of the corresponding existing node is converted to x new For example, x k is x new The existing nodes nearby (at the center of the circle is x new Before this step, x k The parent node is x k-1 In this step, try to k The parent node is transformed into x new , judge x initial to x k Is the total route distance shortened, and x new →x k There is no collision between them. If both of them are satisfied, then x k The parent node is transformed into x new .

[0097] Furthermore, the target is determined. During the expansion of the path tree, the device is constantly checked to see if it is close to the target point, i.e., the x new and x finalIf the distance between them is greater than the set threshold, repeat the steps: random sampling - parent node selection - tree expansion attempt - forward collision check - adding new nodes - parent node reselection - node reconnection; if it is less than the set threshold, terminate the path tree expansion and set x final The parent node position x new , and connect them as the last segment of the path.

[0098] Among them, rotational collision checking is to deal with the situation where the device not only needs to avoid obstacles when encountering them during driving, but also needs to avoid collisions during the device's rotation. In particular, rectangular devices occupy more space when turning than when driving in a straight line, so it is necessary to include rotational collision checking in path planning. The core idea is to use vector cross products to determine whether the device will encounter obstacles during rotation. The rotational collision check includes the following steps:

[0099] S321. Preliminarily determine the collision range based on the spatial intersection between the obstacle and the rotation range of the highest priority device.

[0100] In one embodiment, a preliminary determination is made of the collision range between the obstacle and the highest priority device.

[0101] Specifically, determine whether the obstacle intersects with the device's "rotational collision circle." If no intersection exists, it is determined that there is no collision risk; if an intersection exists, subsequent steps are required to perform a local rotational collision check. The radius of the collision circle is the longest distance from the device's rotation center to its outer boundary. For a rectangular device, the radius is the distance from the center to the corner, as shown in the following formula:

[0102]

[0103] Where r is the radius of the collision circle, a is the width of the rectangular device, and b is the length of the rectangular device.

[0104] S322. Determine a rotation path vector based on the collision range.

[0105] In one embodiment, the device's orientation before rotation (direction vector n1), orientation after rotation (direction vector n2), and the line vector n between the device center and the rotation center are determined based on the current position of the moving device, the position of its parent node, the position of the new node, and the position of the obstacle center. p ,like Figure 2 As shown, where x k is the current location node of the device, x k-1 is x k The parent node, x new is the newly added node, p is the obstacle center, so n1 is determined by x k-1 Point to x k , n2 is determined by xk Point to x new , n p By x k Point to p.

[0106] S323. Based on the rotation path vector, determine the rotation direction and the rotation collision risk.

[0107] In one embodiment, the vector cross product judgment method is used to detect the collision of the walking device in clockwise and counterclockwise rotation. Figure 2 For example, first perform anticlockwise (n1×n2>0) rotation collision detection according to the right-hand rule. If n p ×n2>0 and n p ×n1<0, then the device is at risk of collision during counterclockwise rotation. At this time, the device can avoid collision with the obstacle by rotating clockwise. If the right-hand rule is followed, the clockwise (n1×n2<0) rotation collision detection is first performed. At this time, if n p ×n2<0 and n p ×n1>0, the device faces a collision risk during clockwise rotation. In this case, the device can avoid a rotational collision with the obstacle by rotating counterclockwise.

[0108] Furthermore, we traverse the detection of all stationary obstacles. If the collision detection can be achieved by counterclockwise or clockwise rotation, then x new Can be added to the path tree.

[0109] It's important to note that the rotational collision check mentioned above needs to be supplemented with the method for planning the shortest path for a single device based on the RRT* random sampling concept, as a basis for determining the collision risk of subsequent devices. In this case, the basis for judgment consists of two parts: the travel collision check and the rotational collision check.

[0110] S4. Based on the shortest path, perform a collision check on the travel paths and rotation paths of multiple devices.

[0111] In an unmanned scenario, multiple devices operate simultaneously, so the path planning between devices needs to consider not only collisions with static obstacles, but also dynamic interactions with other devices to ensure that multiple devices do not collide in the same area. There is a time difference between the front and back processes during the construction process. Therefore, the present invention, assuming that the path of the high-priority device (front device) has been determined, focuses on coordinating the paths of the low-priority devices (rear devices) to achieve collision-free interaction of the equipment group. Therefore, the present invention performs collision checks on low-priority devices based on the planned paths of the high-priority devices. The collision check includes a travel path collision check and a travel-rotation collision check.

[0112] In one embodiment, the highest priority device is used as the leading device to perform the path collision check. Figure 3 As shown, device 1 is the highest priority device, and device 2 is a device with one priority lower than the highest priority device, also called the second highest priority device.

[0113] Specifically, it is necessary to determine whether the time scales of the rear device and the front device near the intersection point intersect. Let t 11 to t 12 Represents the time period when the device passes through the intersection area, t 21 to t 22 is the time period during which the device 2 passes through the intersection area. If [t 11 , t 12 ] and [t 21 , t 22 ] If there is an intersection, it is determined that there is a risk of collision between device 2 and device 1. Therefore, in the multi-machine collaborative path planning method of the present invention, it is first determined whether there is a spatial intersection between the development path of the rear device and the path of the front device. If there is no intersection, the path tree is developed according to the normal process. If there is an intersection, it is determined whether there is a temporal intersection between the two in the intersection section. If there is no intersection, the path tree is also developed according to the normal process. If both spatial intersection and temporal intersection exist at the same time, it is determined that a travel path collision will occur and the node of the rear device needs to be abandoned.

[0114] In another embodiment, the highest priority device is also used as the leading device to perform the travel-rotation collision check. Figure 4 As shown, device 1 is the highest priority device, and device 2 is the second highest priority device.

[0115] Specifically, it is necessary to determine whether the rotation process of the rear device at the node and the travel process of the front device at the segment have a time scale intersection. Let t 11 to t 12 Represents the time period when the device passes through the road section, t 21 to t 22 is the time period during which the device 2 rotates at the node, if [t 11 , t 12 ] and [t 21 , t 22 ] there is an intersection, then it is determined that there is a risk of collision between device 2 and device 1. Therefore, in the multi-machine collaborative path planning method of the present invention, for possible travel-rotation collision, first determine whether the rear device is at the path node due to x newWill the rotation of the rear device intersect with the movement of the front device in space? If there is no intersection, the path tree is developed according to the normal process. If there is an intersection, it is determined whether the rotation time period of the rear device and the movement time period of the front device have a time intersection. If there is no intersection, the path tree is also developed according to the normal process. If both spatial intersection and time intersection exist, it is determined that a movement-rotation collision will occur, and the new node x of the rear device needs to be discarded. new .

[0116] In another embodiment, path planning for multiple devices is added based on the pre-planned paths for Device 1 and Device 2. Lower-priority devices perform collision checks on their travel paths, as well as their travel and rotation paths, based on the paths planned for all higher-priority devices. For example, the path for Device 3 is planned based on the path planning results for Device 1 and Device 2. Following the method used to plan the travel and rotation paths for Device 1 and Device 2, traversal planning is performed for all devices.

[0117] It's important to note that multi-machine collaborative dynamic collision checking also needs to be supplemented with the RRT* random sampling approach to plan the shortest path for a single device, serving as a basis for determining the collision risk of subsequent devices. This basis consists of three parts: travel collision checking, rotational collision checking, and multi-machine collaborative dynamic collision checking.

[0118] S5. Based on the results of the collision check, realize the multi-dimensional obstacle avoidance global path collaborative planning of the equipment group.

[0119] In one embodiment, multiple rectangular moving devices are set up, and through moving collision detection, rotation collision detection and multi-machine collaborative dynamic collision detection, multi-dimensional obstacle avoidance global path collaborative planning of the equipment group is realized.

[0120] Specifically, first, a global coordinate system without artificial ground is set, and the equipment movement and point sampling intervals are delineated according to the actual project. In this embodiment, the equipment activity interval is delineated as a square with a length and width of [-2, 18].

[0121] Furthermore, several static obstacles without artificial ground are set, including rectangular obstacles and circular obstacles. The rectangular obstacles are composed of the sequence (x r1 ,x r2 ,a,b) means, where x r1 ,x r2 represents the center coordinate of the obstacle, a is the horizontal length of the obstacle, and b is the vertical length of the obstacle. The circular obstacle is represented by the series (x c1 ,x c2 ,c) means, where x c1 ,x c2represents the center coordinates of the obstacle, and c is the radius of the obstacle. In this embodiment, the rectangular obstacles are (5, 5, 0.25, 0.5), (3, 8, 1, 0.5), and (7, 3, 0.5, 1), and the circular obstacles are (3, 6, 1), (3, 10, 1), (9, 5, 1), and (8, 10, 1).

[0122] Furthermore, the size, starting point, target destination, and departure interval of the traveling device are set. In this example, three rectangular traveling devices are set up, with lengths and widths of 1×0.5, 2×0.5, and 2×0.5, respectively. Their starting points are [0,0], [1,0], and [15,0], respectively; their destinations are [15,12], [14,13], and [10,15], respectively; and the departure intervals of the preceding and following devices are set to 10s and 20s.

[0123] Furthermore, the method of step S3 and step S4 is used to perform global path collaborative planning for multi-element obstacle avoidance of the equipment group, and the path result is as follows: Figure 5 shown.

[0124] See Figure 6 , Figure 6 This is a schematic diagram of the structure of a system for unmanned device group path planning based on the RRT* algorithm in an embodiment of the present invention. The system includes an input device, a processor, an output device, and a memory, wherein the input device, the processor, the output device, and the memory are interconnected. The memory is used to store a computer program, which includes program instructions. The processor is configured to invoke the program instructions. The system uses the unmanned device group path planning method based on the RRT* algorithm.

[0125] In this embodiment, the input device includes a sensor network, a GPS positioning system, and a user input device, which is used to provide the system with necessary environmental information and device status information so that the processor can perform path planning.

[0126] Specifically, the sensor network includes radars and cameras, which are used to perceive the position, shape and size of obstacles on the construction site, as well as the movement status of the equipment group in real time; the GPS positioning system provides precise location information for each device on the construction site so that the system can accurately plan the path; the user input device includes a touch screen, keyboard and mouse, which are used to set the priority influencing parameters of the equipment path planning and other path planning related parameters.

[0127] The processor is the core part of the system, which is used to receive information provided by the input device, run the RRT* algorithm for path planning, and output the results to the output device. It includes a data processing module, a path planning module, an optimization module and a control instruction generation module.

[0128] Specifically, the data processing module receives information provided by the sensor network, GPS positioning system and user input device, and performs data preprocessing and format conversion for subsequent algorithm processing; the path planning module runs the RRT* algorithm to generate one or more feasible paths based on information such as the device's starting position, target position and obstacle distribution; the optimization module optimizes the generated path, including smoothing, local adjustment, etc., to improve the quality and feasibility of the path; the control instruction generation module generates control instructions based on the path planning results and the current status of the device, and sends them to the device through the communication module to achieve automatic navigation and obstacle avoidance of the device.

[0129] The output device is used to display the path planning results generated by the system and send control instructions to the device, including a display screen, a printer and a communication module;

[0130] Specifically, the display screen is used to display the real-time location of the construction site equipment, the planned path, the obstacle distribution and other information, as well as provide system status and error information; the printer is used to print path planning reports, equipment status reports, etc. for management personnel to review; the communication module is used to send the path planning results and control instructions generated by the processor to the equipment on the construction site to realize automatic navigation and obstacle avoidance of the equipment.

[0131] The memory adopts a high-speed solid-state hard disk to store the basic parameters input by the input device and the calculation results after processing by the processor. The memory has the characteristics of fast reading and writing speed, large capacity and high reliability, and can meet the needs of large data storage.

[0132] In summary, the present invention provides a method for path planning of equipment groups in unmanned areas based on the RRT* algorithm. By determining equipment priority, the method reduces equipment travel time and improves overall construction efficiency. By integrating the travel rotation obstacle avoidance strategy and the static-dynamic obstacle avoidance strategy, the method avoids the limitations of only considering travel obstacle avoidance and improves the applicability and accuracy of path planning. The method includes: planning the shortest path of a single device under multiple obstacle conditions according to the RRT* algorithm logic; for the rotation collision problem that the path node of the rectangular walking device may face near the obstacle, a rotation collision check is added to the original path planning process as one of the bases for determining whether it can be included in the path tree node; for the problem of multi-machine collaboration in unmanned areas, a collision interaction check of multiple devices is supplemented on the basis of single-device path planning as one of the bases for determining whether the path sampling node of the post-device can be included in its path tree; outputting the coordinates of the planned path nodes of multiple devices and the rotation posture at the nodes, thereby forming a global planning path that serves the unmanned multi-machine collaboration. The present invention provides a human-free equipment group path planning system based on the RRT* algorithm. Through the efficient search capability of the RRT* algorithm, it can quickly plan a safe and efficient path for the equipment group in a complex construction site environment, effectively avoid collisions and path conflicts between equipment, and ensure the orderly progress of construction site operations. Secondly, the optimization capability of the RRT* algorithm makes the planned path smoother and more continuous, reducing the number of sudden stops and turns of the equipment, thereby improving the operating efficiency and lifespan of the equipment. Finally, the system also has strong adaptability and flexibility, and can be dynamically adjusted and optimized according to different construction site environments and equipment requirements, realizing adaptive and intelligent path planning.

[0133] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some or all of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and description of the present invention.

Claims

1. A method for human-free device group path planning based on RRT* algorithm, characterized in that: The method comprises the following steps: Obtain information on the number, starting point, end point, travel length, mission urgency, energy efficiency and obstacle areas of unmanned equipment; Determine the priority of the devices in the path planning based on the information; Determining the priority ranking of the devices in the path planning according to the information includes: Based on the information, a priority evaluation model of the device in path planning is constructed; The priority evaluation model satisfies the following expression: Q i =αP i +βE i +γT i Among them, Q i is the priority evaluation score of the i-th device, P i is the task urgency of the i-th device, E i is the energy efficiency of the i-th device, T i is the path obstacle influence of the i-th device, α, β, γ are weight coefficients; Among them, M i is the straight-line distance from the starting point to the end point of the i-th device, K i is the total number of obstacles on the straight path from the starting point to the end point of the i-th device, G is the total number of obstacles in the entire unmanned area, R i is the progress of the i-th device; Determining the priority ranking of devices in path planning using the priority evaluation model; Based on the priority ranking, determining the shortest path for the highest priority device under multiple obstacle conditions; Determining the shortest path for the highest priority device under multiple obstacle conditions based on the priority sorting includes: Based on the priority ranking, determining a highest priority device; Planning a travel path for the highest priority device to determine the shortest path under multiple obstacle conditions, wherein the planning includes a travel collision check and a rotational collision check; Planning the travel path of the highest priority device and determining the shortest path under multiple obstacle conditions includes: At the starting point of the highest priority device, generating a root node of the path tree; Random sampling is performed within the travel area of ​​the highest priority device to generate a random point; The shortest distance from the random point to each node of the path tree is determined as the parent node; Generate a new path at the parent node along the direction toward the random point with a length equal to the travel length, wherein the end of the new path is a proposed growth node of the path tree; Comparing the distance between the new path and the obstacle with the size of the traveling equipment to determine whether there is a collision risk; According to a result of no collision risk, the proposed growth node is added to the path tree; Identify existing nodes within a certain range of the node to be grown, and use the point with the shortest total path distance from the starting point to the node to be grown as a new parent node; Determine whether the total path distance of the existing nodes is shortened after reconnecting with the proposed growth node as the parent node, and determine whether there is a collision risk; Performing a threshold judgment on the distance between the proposed growth node and the end point to determine the shortest path under multiple obstacle conditions; Based on the shortest path, performing collision checks on the travel paths and rotation paths of multiple devices; Through the results of the collision check, the multi-dimensional obstacle avoidance global path collaborative planning of the equipment group is realized.

2. The method for non-manual equipment group path planning based on the RRT* algorithm according to claim 1, characterized in that: The rotation collision check comprises the following steps: Preliminarily determining the collision range based on the spatial intersection of the obstacle and the rotation range of the highest priority device; determining a rotation path vector based on the collision range; Based on the rotation path vector, the rotation direction and the rotation collision risk are determined.

3. The method for non-manual equipment group path planning based on the RRT* algorithm according to claim 2, characterized in that: Based on the rotation path vector, determining the rotation direction and the rotation collision risk using a vector cross product method includes: Determining a counterclockwise rotation collision risk based on the rotation path vector; Based on the rotation path vector, the clockwise rotation collision risk is determined.

4. The method for non-manual equipment group path planning based on the RRT* algorithm according to claim 1, characterized in that: The performing of collision check on the travel paths and rotation paths of multiple devices based on the shortest path includes: Based on the shortest path of the highest priority device, the travel path of the lower priority device is checked for collisions; The travel and rotation paths of lower priority devices are checked for collision based on the shortest path of the highest priority device.

5. The method for non-manual equipment group path planning based on the RRT* algorithm according to claim 1, characterized in that: The collision check includes: The low-priority device performs collision checks on its travel path based on the paths planned by all higher-priority devices; The lower priority device performs collision checking of rotation and travel paths based on the paths planned by all higher priority devices.

6. A system for planning a path for a group of equipment without human intervention based on an RRT* algorithm, wherein the system uses a method for planning a path for a group of equipment without human intervention based on an RRT* algorithm according to any one of claims 1 to 5, characterized in that: The system includes an input device, a processor, an output device and a memory, wherein the input device, the processor, the output device and the memory are connected to each other, wherein the memory is used to store a computer program, the computer program includes program instructions, and the processor is configured to call the program instructions.

Images